CD4 lymphocyte levels have been shown to correlate well with stage of HIV-1 infection, but their predictive value for future rates of CD4 decline or for time to development of AIDS is much poorer. We propose here that station measurements of CD4 levels do not take into account the dynamics of T cell loss and replacement. We will therefore, refine currently available methods for identifying newly generated T cells and use them to assess T cell turnover rates in HIV-1 infected individuals, much as reticulocyte counts are used to evaluate erythrocyte turnover. We will test the hypothesis that individuals with similar levels of circulating T cell subsets may be at very different risk for AIDS, depending on the extent of HIV-induced lymphocyte replacement. We hypothesize that the increased prevalence of T cells bearing surface determinants that are conventionally interpreted as memory T cell activation markers in HIV-1 infection represents an accumulation of immature T cells rather than the activation of memory cells. Despite the phenotypic similarity between the two developmental stages, as shown in our preliminary data as well as data from other laboratories, activated memory cells would be immunologically competent whereas immature cells would not. Criteria used to distinguish immature from activated T cells will be confirmed on the basis of phenotypic and functional profiles expressed by cells obtained from two types of patients known to be generating large numbers of new lymphocytes, namely bone marrow transplant recipients and neonates. The characteristics of newly generated lymphocytes in these individuals will be compared with those of circulating T lymphocytes from HIV-1-seropositive and - seronegative homosexual men being followed in the Baltimore center of the Multicenter AIDS Cohort Study (MACS). This information will provide an initial validation of our immature T cell phenotypic profile in HIV-1 infected individuals. Next, the different T cell maturational components will be quantified longitudinally in selected members of the Baltimore cohort before and at intervals after seroconversion. Fresh and cryopreserved specimens from individuals of known clinical outcome will be tested. The data will be used to construct a mathematical model for assessing HIV-1 disease progression based on the maturational composition of T cells in individuals with similar CD4 cell counts (i.e., the estimation of the effect of CD4 cell immaturity after controlling for the absolute CD4 cell number). These analyses will be conducted in seroprevalent and seroconverter members of the Baltimore cohort using techniques which we have previously applied to lymphocyte population shifts in MACS. The data obtained may add to our understanding of the natural history and pathogenesis of HIV-1 infection.